INCLUSION
PROBABILITY PROPORTIONAL TO SIZE
-Prakhar Srivastava
INTRODUCTION:
Sample Survey Design is an integral aspect of
Statistical Analysis. While designing a sample survey, it is also equally
important to choose a suitable variance estimation method.
The sampling design is created by constructing a
sample space in which the population units occur in proportion to their sizes
and every pair of units occur in such a way as to provide a non-negative and
stable variance estimator. The selection procedure is then straightforward and
the values of the inclusion probabilities of pairs of units can be computed
easily.
IMPORTANCE OF INCLUSION PROBABILITY:
1.
Knowing the inclusion probability of each element in the
population is very important for sampling theory. Inclusion probabilities give
an insight into how the RSS design has more control over which element enters
the sample than SRS has.
2.
In probabilistic sampling, each element of the population must
have a non-zero probability of being included in a sample, otherwise unbiased
estimation is not possible.
3.
The reason, therefore, is straightforward: If not all elements of the population have a positive probability of becoming part of a sample, one cannot
expect that an actual sample can describe the unknown population
parameters correctly.
4.
The inclusion probability πi refers to the chance that the
ith population element becomes part of a sample.
5.
The inclusion
probability should be distinguished from the selection probability p(s) of a sample that is the probability that a certain unordered set of elements (e.g., a
number of trees included by a sample plot) is selected as a sample.
IMPORTANT FORMULAE FOR INCLUSION PROBABILITY:
1. Horvitz and Thompson
2. Yates
Grundy
PROPERTIES OF INCLUSION PROBABILITY:
APPLICATION IN R
Computes the first-order inclusion probability of each unit in the population given a fixed sample size design
Thus, this example gives proper inclusion of each
variable, and with a proper sum of inclusion as 2.
REFERENCES
1. IPPS
Sampling Methods
https://www.researchgate.net/publication/277259216_A_class_of_ipps_sampling_schemes
2. Pik:
Inclusion Probabilities for Fixed Size Without Replacement
https://rdrr.io/cran/TeachingSampling/man/Pik.html
Comments
Post a Comment